A Bayesian, Nonlinear Particle Filtering Approach for Tracking the State of Terrorist Operations

In this paper, we describe a novel approach to track the progress of suspected terrorist operations and to optimize courses of action to delay or disrupt these operations. The approach uses Monte Carlo sampling and Bayesian, nonlinear particle filtering to estimate the state (schedule) of a terrorist operation. The operation is specified in the form of a project management model (such as a Program Evaluation and Review Technique (PERT) model) with uncertain task durations. We describe the underlying algorithms for performing the estimation given a set of observables of variable quality, and evaluate the effectiveness of the techniques through a series of numerical experiments that include a wide range of data characteristics.